fisheries, ecosystems, and aquaculture

FISHERIES, ECOSYSTEMS, AND AQUACULTURE 33. ENVIRONMENTAL EFFECTS OF SALMON AND MUSSEL AQUACULTURE Barry T. Hargrave*, David J. Wildish, and Peter J. Cranford Summary

Culture of salmon in floating net-pens in the Western Isles region of the Bay of Fundy and of blue mussels on suspended long-lines in Prince Edward Island expanded between 1980 and 1990. Both types of aquaculture can potentially cause adverse environmental changes in water and sediments where depth is shallow and tidal exchange with offshore waters is restricted. High fish biomass, metabolism, and feed pellet deposition may cause nutrient enrichment during salmon cage culture if the capacity of an inlet to assimilate additional organic matter is exceeded. Mussels filter natural suspended matter, but the release of faecal matter increases sedimentation and the potential for benthic enrichment. Starting in the 1990s, research staff from the Bedford Institute of Oceanography, St. Andrews Biological Station, provincial governments, and university laboratories collaborated to examine farm- to ecosystem-scale environmental effects associated with intensive salmon and mussel culture. Physical and bioenergetic models were developed to describe local and inlet-wide effects such as dissolved oxygen and suspendedmatter depletion, increased sedimentation, and changes in benthic communities and habitats in areas where salmon and mussel farms were concentrated. The research provided environmental managers and provincial licensing authorities with information to identify local and farfield environmental problems potentially due to aquaculture and suggested approaches to mitigate effects once changes occur. The aim was to allow salmon and shellfish aquaculture industries to expand while operating in an environmentally sustainable manner. Key words: mariculture, production, ecological carrying capacity, ecosystem impacts. *E-mail: [email protected]

Research on factors controlling growth in bivalves such as the Blue Mussel (Mytilus edulis) and Common Oysters (Crassostrea virginica) was carried out in the 1950s and 1960s in southwestern New Brunswick (SWNB) and at the Fisheries and Oceans Canada (DFO) field station at Ellerslie, Prince Edward Island. The research was focussed on establishing optimal conditions for oyster and mussel culture and in addition considered habitat, morphological variations, and genetic diversity effects on growth and diseases in natural populations. Research on Atlantic Salmon (Salmo salar) in the 1970s was focussed primarily on nutritional requirements and husbandry practices to optimize growth and disease protection (Anderson 2007). Lack of knowledge about the control and treatment of aquatic diseases was a limiting factor impeding the growth of salmon and oyster aquaculture industries at the time. As knowledge about the control and treatment of aquatic diseases and the development of optimal conditions for salmon nutrition increased, the marine environment was considered but only in terms of how physical conditions affected growth and tissue quality of wild and cultured species. Few studies considered the converse: i.e., how the growth of cultured animals affects the environment. An associated question was: are there ways to determine the optimum holding capacity (for salmon) or production carrying capacity (for molluscs) in an inlet or bay that allows production to be optimized while preventing adverse environmental changes? Research programs initiated in the 1980s within the DFO Science Branch at Bedford Institute of Oceanography (BIO) in collaboration with staff at St. Andrews Biological Station (SABS), St. Andrews, New Brunswick, were designed to address these questions (Fig. 1). The research assisted habitat managers and provincial fisheries and aquaculture authorities with licensing and regulatory responsibilities for these new and expanding industries. With the exception of Prince Edward Island, where DFO has regulatory authority for site approval, licensing of aquaculture leases in Canada is controlled by the fisheries and aquaculture departments in provincial governments. The mandate during the 1980s was to expand aquaculture industries as fast as possible to provide employment and economic development, particularly in rural areas where incomes from traditional harvest fisheries were in decline. The hope was that aquaculture production could replace some of the losses in traditional harvest fisheries which had markedly declined in eastern Canada between mid-1960 and 1980. Cultivation methods for various invertebrate and finfish species were being investigated around the world in the 1970s based on advances in developing broodstock and rearing of juveniles for stocking purposes. Environmental conditions in some coastal areas of eastern Canada such as SWNB were especially favourable for salmon aquaculture. Farms were

Figure 1. Map showing locations of sites in the Atlantic Provinces mentioned in the text (base map from Shaw et al. 2006).

established where physical conditions such as temperature, salinity, and protection from excessive wave action and high currents allowed salmon culture. In the case of Blue Mussels, natural suspended particulate matter (seston) in the water is the food supply, so culture conditions are most favourable in inlets where water exchange provides a continuous supply of particles suitable for ingestion. Finfish and shellfish aquaculture can have both positive and negative environmental consequences (Fig. 2). The major difference is that growth of salmon in floating net-pens requires addition of food pellets which if uneaten along with fecal matter settle to the bottom. Potential adverse effects of salmon aquaculture were reviewed by Fisheries and Oceans Canada (2003, 2004, 2006). High fish biomass creates opportunities for viral and bacterial diseases such as Bacterial Kidney Disease, Furunculosis, and Infectious Salmon Anaemia or parasitic infections to develop which are often controlled by application of therapeutants and vaccines. Antibiotics and other drugs used for disease treatment can have potentially negative effects on non-target species. In addition, naturally occurring bacterial populations can develop resistance to antibiotics added to fish feed. Escapement of salmon and resulting negative interactions with wild stocks is another important issue occurring whenever net pens are located close to migratory routes of natural stocks. Mussels feed on naturally available suspended particulate matter (seston) so no organic matter (OM) or nutrients are added. The potential

§

§

Voyage of Discovery

Fifty Years of Marine Research at Canada’s Bedford Institute of Oceanography 1962 - 2012 Bedford Institute of Oceanography-Oceans Association, Dartmouth, Nova Scotia, Canada (2014)

page 285

FISHERIES, ECOSYSTEMS AND AQUACULTURE

A

(Cranford et al. 2006). The major physical, chemical, and biological factors affecting marine finfish and mussel culture have been well known since the 1990s. Adequate depth and water movement within farms are required to supply dissolved oxygen (DO) and disperse/dilute excretory products and, in the case of mussels, sufficient concentrations of phytoplankton and suspended particles. However, the physical structure of salmon cages and mussel long-lines can interfere with water circulation. Oxygen availability may be reduced within salmon pens due to fish respiration that exceeds DO supply during periods of slack water (Page et al. 2005). The addition of salmon feed can also increase sedimentation under cages and the oxygen demands of deposited OM in feed pellets and faeces which may result in hypoxic (low oxygen) or anoxic (no oxygen) surface sediments (Hargrave et al. 1993, 1997). The changes in sediments due to oxygen depletion may limit growth or pose health risks to the cultured stock as well as create deleterious effects on wild populations of fish, shellfish, and other invertebrates. In addition to local effects under salmon pens or around long-lines, far-field changes distant from farms or on an inlet-wide scale may arise from nutrient enrichment (i.e., eutrophication). Release of excretory products, increased sedimentation, OM enrichment, and accumulation of fine particles in sediments and changes in planktonic and benthic food webs can significantly alter species composition in the water column and benthic communities. Many of these factors were discussed in the Fisheries and Oceans Canada (2003, 2004, 2006) science reviews where potential negative local and far-field environmental effects due to culture operations were summarized. An International Council for the Exploration of the Seas (ICES) workshop held at BIO in September 1995 (Silvert and Hargrave 1995) concluded that sustainable development of aquaculture in the coastal zone required integration of models of environmental impacts with application of tools such as expert systems to assist with management decisions.

B

Research Programs and Funding Sources Figure 2. Principal pathways of dissolved and particulate waste products from: (A) finfish, and (B) shellfish aquaculture (after Cranford et al. 2006: Figs. 3.2 and 3.3). Shaded areas indicate deposition of settled particulate material on the bottom generally concentrated directly under fish pens and mussel long-lines. Distances and horizontal directions of sediment accumulation are affected by the size, number, and configuration of cage and culture line arrays and local variations in currents and bottom topography.

environmental effects of mussel culture are related in large part to how the cultured population interacts with the ecosystem by means of removing particles during suspension feeding. Food depletion by extensive mussel stocking ultimately limits the maximum aquaculture yield (i.e., production carrying capacity) in the water body. Other changes in ecological processes, species, populations, or communities will accompany seston depletion by mussels and knowledge of these effects is required to assess the ‘ecological’ carrying capacity of a region. Changes in ecosystem processes and structure from the byproducts of suspension feeding, including ammonia excretion and faeces production were reviewed in Fisheries and Oceans Canada (2003) and Cranford et al. (2006). The latter increases nutrients and OM in sediments leading to enhanced benthic organic enrichment. When the stock is harvested, OM and nutrients are removed in mussel tissue which could potentially decrease eutrophication caused by nutrient enrichment from other sources (e.g., agricultural runoff, discharges from fish processing plants, etc.). Mussels on suspended long-lines provide a large area of solid surfaces in the water column which may be colonized by attached invertebrates such as tunicates. These filter feeders consume a similar size range of suspended material as food, as do mussels, and their feeding activity can reduce mussel growth and increase rates of particle removal from the water column. Evidence from observations in Prince Edward Island inlets shows that far-field (inlet-wide) changes in the trophic balance between autotrophs (phytoplankton) and heterotrophs (bacteria) may occur as a result of the selective filtration of particles by mussels (Harrison et al. 2005, Kepkay et al. 2005). Similarly, the limited capacity of mussels to filter small particles can result in a shift towards smaller phytoplankton species dominating the base of the food chain

§

Voyage of Discovery

§

page 286

As salmon and mussel aquaculture industries expanded in New Brunswick, British Columbia, and to some degree Quebec and Newfoundland after 1980, DFO funding was available for research on aquaculture-environment interactions. Projects by Wildish et al. (1988, 1990) compared water column and sediment properties proximate to and distant from salmon farms to determine if environmental changes could be detected. Salmon farms were highly concentrated in specific areas (L’Etang Inlet, Lime Kiln Bay, and Bliss Harbour) and research was focussed in these locations (Fig. 3). The Coastal Oceanography for Sustainable Aquaculture (COSAD) program was also formed within the Ocean and Ecosystem Science Branch of DFO at the same time with the aim of supporting physical and biological observations in coastal areas where salmon and mussel culture facilities were expanding (Page et al. 1999). In 2000, the Environmental Sciences Strategic Research Fund (ESSRF) supported scientists from BIO, SABS and local universities to collaborate in expanding aquaculture-environment interactions research. Environmental Studies for Sustainable Aquaculture (ESSA) was a threeyear (2000-03) ESSRF-supported project to identify measures and model far-field effects of salmon aquaculture with study sites in SWNB, Newfoundland and British Columbia. Although SWNB had been a focus for the initial research on environmental effects of salmon aquaculture on water and sediments, projects in Bay d’Espoir, Newfoundland, and the Broughton Archipelago, British Columbia, co-ordinated by R. Anderson and T. Sutherland, respectively, contributed significantly to the national scope of the project (Anderson et al. 2005, Stucchi et al. 2005). Prior studies had focussed on local impacts on, under and in close proximity to net-pens, but physical circulation models describing the distribution of DO, nutrients, and suspended particulate matter developed during the ESSA project provided a framework for estimating the dispersion of dissolved and particulate waste from farms (Hargrave 2004). The Aquaculture Collaborative Research and Development Program (ACRDP) was also initiated within DFO in 2000. ACRDP projects required industry partnerships that provided up to 30% of the funding with the aim of supporting studies to increase the competitiveness of the Canadian aquaculture industry. Projects had to show benefits

Hargrave et al. — Aquaculture Environmental Effects

Figure 4. Harvesting mussels in Tracadie Bay, Prince Edward Island, July 2002. Figure 3. Photograph looking east-northeast across Lime Kiln Bay, New Brunswick, prior to 1997 showing locations of salmon net pens; L’Etang Inlet is visible in the distance (photo: K. Haya/J. Martin, DFO-SABS, NB).

to the industry, and it proved difficult to establish partnerships and obtain funding for environmental research that might show negative effects which could ultimately limit expansion of culture activities. The COSAD program supported studies of mussel-environment interactions to develop physical models describing water exchange in Prince Edward Island inlets where intensive mussel aquaculture was established. The first field studies occurred in Tracadie Bay, an inlet intensively studied over the next decade to measure environmental effects of intensive mussel culture (Page et al. 1999). Research on factors affecting bivalve growth and environmental interactions was also enabled by the establishment of research networks within the Natural Sciences and Engineering Research Council (NSERC) in 1990. NSERC supports university-based research, but collaboration with government and industry laboratories is encouraged. DFO staff from BIO participated in the Ocean Production Enhancement Network (OPEN) from 1990 to 1994, one of NSERC’s first research networks. The multidisciplinary study identified factors controlling fish and shellfish growth and reproduction. It provided opportunities for graduate students, and university and government researchers to investigate growth and physiological responses of bivalves (mussels and scallops) to environmental changes and indirectly to environmental effects caused by culture activities. While OPEN focussed on Atlantic Cod (Gadus morhua) and scallops, factors important for bivalve growth in coastal waters in general were investigated by assessing water column and benthic effects of mussel and scallop culture in Upper South Cove near Lunenburg, Nova Scotia (O’Dor and Thompson 1996). DFO-based studies initiated during OPEN continued after 2000 supported by an ESSRF project (Integrated Ecosystem Studies for Modelling Mussel Aquaculture-Environment Interactions) with collaboration between research scientists and biologists from DFO (Maritimes, Gulf and Quebec regions), Dalhousie University (Halifax, NS), and IFREMER (Institut français de recherche pour l’exploitation de la mer, France). Results from investigations of the consequences of mussel culture to ecosystem structure and function were incorporated into numerical models to benefit from their predictive power. The project was notable at the time for taking a multidisciplinary ecosystem approach that utilized the broad oceanographic capabilities of BIO scientists. Modelling activities combining research in physical, chemical, and biological oceanography addressed broad-scale questions regarding system productive capacity, food depletion, nutrient cycling, and aquaculture/land-use interactions. Research was primarily located in Tracadie Bay, the most extensively leased mussel aquaculture embayment in Canada (Fig. 4), but mussel culture effects on ecosystem processes were also assessed in other Prince Edward Island inlets (Cranford et al. 2006, 2007, 2009; Grant et al. 2008; Hargrave et al. 2008). The Program for Aquaculture Regulatory Research (PARR) was established in 2008 to fund DFO projects seeking new knowledge to support and advise ecosystem-based environmental regulations and decision making related to aquaculture. PARR funded an assessment of the spatial and temporal scales of environmental effects of mussel aquaculture in St. Anns Harbour, Nova Scotia, the location of the largest

mussel lease application approved (2003) in the Maritimes. Scientists from BIO and Dalhousie University collaborated with Nova Scotia Fisheries and Aquaculture to test the effectiveness of environmental impact assessment predictions and assess the environmental monitoring program (EMP). In 2008 an intensive synoptic survey was conducted to test the effectiveness of the EMP for detecting benthic changes due to mussel culture operations. The project documented temporal and spatial changes in benthic and pelagic performance indicators, provided an assessment of the effectiveness of the current EMP, and made recommendations for improving monitoring methods (Cranford et al. 2011).

Major Research Results Local Versus Far-field Effects Studies

DFO research on aquaculture-environment interactions in the Maritimes described above focussed on studies of local effects within salmon and mussel farm boundaries. Increased OM sedimentation, sedimentwater exchange of DO and ammonia, and altered biomass and species composition of benthic invertebrates (macrofauna) were observed in some locations under and adjacent to salmon net-pens and mussel lines but effects decreased with distance (Fisheries and Oceans Canada 2003). Depending on water depth and current speed, measureable differences from reference (non-farm) sites were restricted to within 50 m of culture facilities. Changes in DO and ammonia in the water column within a salmon farm leased area were more difficult to detect. DO varied with tidal currents, but more importantly, with the reduced tidal volumes associated with low water during a tidal cycle. If changes in concentrations were detected, they only occurred briefly when water currents were low (Wildish at al. 1993). The observations were consistent with model calculations for oxygen budgets in SWNB (Trites and Petrie 1995, Strain et al. 1995, Page et al. 2005) where fish respiration and water circulation were concluded to be major factors controlling DO within salmon cages. Local effects of salmon net-pen culture on sediments measured in SWNB in the mid- to late 1990s (Hargrave et al. 1997) suggested that it might be possible to monitor environmental effects of farms using observations at individual sites. Comparisons of benthic biological and geochemical variables at different salmon leases in SWNB showed that total dissolved free sulfides (S) and oxidation-reduction potentials (Eh) in the surface (upper 2 cm) sediment were sensitive to excessive OM sedimentation. These variables were recommended as the basis for monitoring programs in an EMP (Wildish et al. 1999). There was widespread interest in this approach by regulatory agencies since it would make lease owners responsible for monitoring their farm site. Licensing could require that an EMP be applied by an operator within their leased area with results reported annually to provincial regulatory agencies. The review that examined potentially negative environmental effects of salmon net-pen and mussel long-line culture over areas larger than individual farms or lease areas (Fisheries and Oceans Canada 2003) emphasized the lack of far-field environmental effects studies. Observations at the scale of a single farm or within a lease that

§

§

Voyage of Discovery

page 287

FISHERIES, ECOSYSTEMS AND AQUACULTURE described local environmental conditions could not be used to determine the capacity of an inlet or bay to assimilate additional dissolved and particulate nutrients released from culture operations. The examples below show how field observations and model calculations were combined in attempts to consider inlet-wide impacts of multiple salmon or mussel farm leases.

Evidence for Far-field Effects Due to Salmon Culture in the Bay of Fundy

The first evidence for far-field effects of salmon culture in SWNB was provided by Pohle et al. (2001) in a study of benthic macrofauna communities and other sediment variables. The temporal study (199499) with annual sampling was located in areas of SWNB where intensive salmon culture had developed with sites at least 200 m away from operational farms. Significant changes in benthic community structure were noted in the first year (1994-95), coincident with an increase in salmon culture production, which persisted for the remainder of the study. A second study (Wildish and Pohle 2005) compared macrofauna data collected from the same area before salmon culture had begun (197475) with similar data collected in 1997 and 2000 after the industry was well developed. Macrofauna community structure at a reference station in the St. Croix Estuary in SWNB covering 10 km had not changed during this period showing that no general, long-term changes in biomass or taxonomic composition had occurred during the study period. A characteristic feature of benthic macrofauna communities in the lower L’Etang in SWNB in 1975 was the presence of extensive beds of burrowing, filter-feeding amphipods. Increased sedimentation of flocculated fine-grained sediment throughout the inlet during the 1990s is thought to have caused the disappearance of these communities (Milligan and Law 2005). By 1997 to 2000, amphipod beds, historically an important food resource for young groundfish such as haddock and cod, had completely disappeared from Lime kiln Bay. An independent study by Robinson et al. (2005) demonstrated that macroalgal eutrophication had occurred on local beaches in SWNB due to nutrients released from salmon farms. Intertidal areas near salmon farms had high standing crops of the green alga Ulva spp. that was found to reduce recruitment, growth, and survival, and cause behavioural changes in commercially important Soft-shelled Clam (Mya arenaria) populations. Circumstantial evidence, in the form of increased levels of zinc in sediments (derived from high levels in salmon feed), suggested that increased Ulva biomass was due to nutrients released from the nearby salmon farm. Yeats et al. (2005) showed that depending on mineralogy and grain size, elevated Zn and Cu, normalized for lithium, could be detected above concentrations normally present in sediments at some distance from salmon cage sites in depositional areas of SWNB.

Numerical and Finite Element Circulation Modelling

The linkage between water circulation and holding capacity for salmon management in the L’Etang Inlet in SWNB was first investigated by Loucks (1988, 1991) and Trites (1991) using a numerical model (de Margerie et al. 1990) described in Wildish et al. (1990). Measurements over two months (May to December in 1989) showed complex horizontal currents with a net inward flow. Greenberg et al. (2005) described the tidally-driven circulation in Passamaquoddy Bay and inlets within SWNB in more detail using a finite element model. Trajectories of soluble material or neutrally buoyant particles added at different locations could be followed hourly over 2.5 tidal cycles. The model output clearly showed large horizontal exchanges in the L’Etang area. Tidally-driven water movement, as in many locations favourable for salmon culture, can transport dissolved substances and suspended particles over long distances between adjacent inlets over a few tidal cycles. The particletracking feature of the model could be used to examine the spread of disease, to determine minimum currents required to replenish oxygen consumed by fish within cage arrays, and to predict sedimentation patterns around farms. The modelled trajectories of water and suspended particles show that management practices in one bay should take into account culture operations in adjacent inlets.

Mass Balance Models

Several studies have used a mass balance approach to compare physical and biological processes of DO supply and loss on an inlet-wide basis to

Nitrogen Fluxes in Tracadie Bay

Similarly to finfish culture, mussel aquaculture influences some fundamental ecosystem processes with a complexity of interactions. Particle depletion by filter feeding and increased biodeposition are key processes with potentially negative ecological effects at inlet-wide scales (Cranford et al. 2006, 2007, 2009; Grant et al. 2008) and models are essential for predicting these effects. A nitrogen budget and dynamic water exchange model developed for Tracadie Bay, Prince Edward Island, showed that most nitrogen flux for the entire inlet occurs as a result of tidal exchange (Cranford et al. 2007). More nitrogen leaves via offshore tidal exchange than enters as river input. The mussel harvest also removes nitrogen from the system, but the amount is small compared to input from river and agricultural sources. The mussels can contribute significantly to the retention of nutrients in the coastal zone, the amounts excreted, sedimented, and buried in sediments. Mussel biodeposition can significantly increase nitrogen flux and concentrations of OM in sediments in shallow inlets with limited water exchange such as Tracadie Bay. Particulate nitrogen fluxes within the bay are dominated by mussel ingestion of suspended material and faecal matter production. Calculations show that benthic enrichment can be expected when biodeposits are buried.

Mapping Inlet-wide Effects of Mussel Culture

Seston depletion and sediment OM enrichment as a result of increased mussel biodeposition can extend over areas much larger than lease boundaries. Eutrophication effects were appearing in sediments in Prince Edward Island inlets as intensive mussel culture expanded (Cranford et al. 2009), and Tracadie Bay served as a suitable case study to evaluate inlet-wide effects. Key sediment variables (water content, grain size, OM, Eh, and dissolved sulfides) measured throughout the inlet showed spatial variations within and outside of lease areas (Cranford et al. 2007, Hargrave et al. 2008). The geochemical variables were used to group stations into benthic enrichment categories that separated mussel lease and non-lease areas. The enrichment gradient was similar to changes occurring as a result of high OM sedimentation near salmon pens described by Hargrave et al. (1993). Shallow water depths, low currents, and a large biomass of cultured mussels in an inlet are required for sedimentation and accumulation of biodeposits to reach the high levels of benthic enrichment observed near salmon pens. Detection of the magnitude and zone of food depletion in and around mussel aquaculture sites is complicated by the large degree of natural variation in phytoplankton and seston concentrations in coastal waters. Two new approaches were developed at BIO to document the scale of phytoplankton depletion. The first was based on conducting rapid highresolution three-dimensional surveys of phytoplankton and suspended particles with a towed undulating sensor vehicle (Cranford et al. 2006,

§

Voyage of Discovery

§

page 288

assess the relative importance of fish metabolism to other processes in SWNB. Page et al. (2005) summarized DO dynamics to show that fish respiration could be an important process causing oxygen depletion within salmon net-pens. Wildish et al. (1990), Trites and Petrie (1995), and Strain and Hargrave (2005) made similar calculations showing that if salmon biomass was sufficiently high, localized DO reductions could occur in areas where currents are weak or water residence times are long. Anecdotal information from some farms in such areas confirmed model predictions that localized oxygen depletion had occurred within fish pens at some locations (Page et al. 2005). The fate of nitrogen in dissolved and particulate wastes released from all salmon farm sites in SWNB in 2002 was also described using mass balance calculations (Strain and Hargrave 2005). A fish-growth model to estimate particulate and dissolved nitrogen loss from feed application and salmon metabolism during a 2-year grow-out cycle showed that most nitrogen release occurs as dissolved ammonia. Faeces production represents a much smaller loss (~10%), but as the particulate matter is almost neutrally buoyant it is potentially available for transport over relatively long distances. While heavier waste feed pellets are buried locally, more buoyant faecal matter can be transported over relatively long distances to be either decomposed in the water column or buried. The mass balance calculations compared nitrogen fluxes through salmon farms with those due to natural processes to show that, where farms were most concentrated, fluxes due to salmon could be increased by more than 300%.

Hargrave et al. — Aquaculture Environmental Effects Grant et al. 2008). This approach has permitted the routine mapping of depletion zones in many mussel aquaculture regions in Canada and Europe. The second approach specifically targets ecosystem-scale effects of phytoplankton depletion by measuring changes in phytoplankton sizestructure resulting from the size-selective feeding behaviour of mussels. Size-fractioned chlorophyll a analysis was shown to be an ecologically relevant and inexpensive indicator of bay-scale changes at the base of the pelagic food web. Picophytoplankton (0.2 to 2.0 µm size range) were shown to dominate in Prince Edward Island embayments where feeding by cultured mussels removes food particles faster than they can be replaced by tidal exchange (Harrison et al. 2005).

DFO research on aquaculture-environmental interactions has been and continues to be focussed on providing advice for habitat managers to support development of sustainable salmon and mussel culture industries. Research results have been used to recommend management approaches to avoid a Harmful Alteration, Disruption, or Destruction (HADD) of marine fish habitat as required under the federal Fisheries Act. Methodologies for habitat impact assessment resulting from the research over the past two decades have become part of a standardized EMP for determining benthic enrichment effects associated with salmon and mussel aquaculture (Wildish et al. 1990, 1999; Cranford et al. 2006). A Canadian Science Advisory Secretariat (CSAS) research document prepared for DFO Habitat Management (Wildish et al. 2005) recommended monitoring methods and identified thresholds for variables shown to be sensitive indicators of benthic organic enrichment (free sulfides and redox potentials) in surface sediments (Hargrave et al. 1997) to assess environmental effects of salmon aquaculture. The methods and assessment approaches are currently applied in an annual EMP at all licensed aquaculture sites in Newfoundland, Nova Scotia, New Brunswick, and British Columbia. Decision support tools were also developed to assist managers with regulatory responsibility (Silvert 1994). A decision support system (DSS) described by Hargrave (2002) consisted of questions about far- and near-field variables potentially affected by salmon aquaculture. Input data for farm variables (the number and sizes of pens and fish stocking density) were combined with physical (mean water depth, presence or absence of sills, tidal amplitude and variations in current speed) and sediment variables (grain size, organic content, S and redox potentials). Scoring based on input data was used to provide a score that reflected the degree that environmental conditions might be adversely affected by the increased organic input due to the presence of the farm. This DSS has been used by DFO Habitat Management staff in the Maritimes Region since 2003 to screen new applications to determine if a potential exists for a proposed farm site to create conditions for a HADD. The CSAS research document of Cranford et al. (2006) recommended a tiered approach for assessing environmental risks of mussel aquaculture where, depending on previous observations and the type of environment and operational nature of the shellfish lease, different levels of monitoring would be applied. In addition, shellfish aquaculture management approaches, developed in part from DFO research, have been recommended to address the needs of worldwide legislations requiring an ecosystem-based management framework (Cranford et al. 2012). Benthic performance indicators and classifications have been implemented as part of global sustainability standards for the certification of cultured shellfish (World Wildlife Fund 2010). Research to improve monitoring methods and develop new approaches continues as both salmon and mussel aquaculture industries wish to expand existing leases and locate new farm sites.

be sensitive indicators of benthic enrichment. Increased sedimentation of fecal matter and waste feed was associated with decreased benthic macrofauna biomass and changes in community structure and the formation of hypoxic or anoxic sediments under and close to net pen and long line arrays. In the past decade these geochemical variables have been applied in provincial EMP programs in Newfoundland, Nova Scotia, New Brunswick, and British Columbia to assess benthic enrichment effects at farm sites due to salmon and mussel aquaculture. While these studies provided information on local benthic enrichment at individual farm sites, they did not allow conclusions about far-field or inlet-wide effects. Numerous field programs (COSAD, OPEN, ESSA, and more recently PARR-funded projects) combined physical, chemical, and biological oceanographic observations and the application of circulation models to determine if effects of salmon and mussel culture could be detected at scales larger than a farm lease. The exchange of water between adjacent inlets in areas of intensive salmon aquaculture in SWNB was demonstrated by a finite element physical mixing model driven by tidal circulation. Model results support a bay management, rather than a farm-by-farm approach for setting stocking densities in areas where many leases are located in close proximity. Mass balance models comparing sources and sinks for DO, particulate carbon and dissolved nitrogen in SWNB where salmon farms are concentrated also showed that chemical fluxes due to cultured salmon could exceed natural fluxes by several times. These observations led to the recommendation that stocking density should be limited in some areas to avoid long-term environmental degradation. A major strength of both near and far-field aquacultureenvironment effects studies has been the multidisciplinary nature of the research. The collaboration of physical, chemical, and biological oceanographers was made much easier because research staff in different disciplines were co-located. In a similar manner, having DFO Habitat Management staff located at BIO allowed research projects to be designed to answer specific, practical questions with results more likely to be used by managers. The development of DSS tools for assessing site suitability for new salmon farms and the tiered approach for EMPs monitoring of mussel farm sites have assisted federal government habitat managers, provincial aquaculture regulators, and salmon and mussel farm operators who apply the site monitoring methods. In addition, DFO research staff have been involved in some EMP monitoring studies which assisted both farm operators and provincial regulators in interpreting the results. Future studies concerning aquaculture-environment interactions must focus on large-scale and long-term effects, impacts on hard bottom areas, and, in the case of finfish aquaculture, the location of farms in deeper water at offshore sites. The geochemical methods currently used in EMPs for assessing aquaculture benthic impacts are only suitable for mud and sand bottom areas where samples can be collected by cores or grabs. Wildish et al. (1999) emphasized that research on monitoring methods should be continued and updated annually. Quantitative methods to assess enrichment effects on rock and cobble substrates and in deeper water must also be developed. Short-term (e.g., annual) and local measures of effects within or close to leases will continue to be carried out by individual aquaculture site operators and companies holding leases as part of ongoing EMP requirements. It is unclear, however, who has the responsibility to consider larger-scale effects beyond lease boundaries that could have broad-scale habitat impacts or effects on non-commercial populations of invertebrates and fish. In the past, aquaculture industries for salmon and mussel culture have not been willing to collect data outside of their lease boundaries. However, DFO does not have the mandate and has limited resources to conduct far-field monitoring studies. Despite the current reduction in research budgets and staffing limitations, research to find practical methods to observe far-field effects must continue. In addition, collaborative management approaches must be developed that consider both local and far-field environmental effects of the growing aquaculture industries.

Synthesis and Future Research

Acknowledgements

Our understanding of near-field environmental effects of salmon and mussel culture based on studies in SWNB and Prince Edward Island increased rapidly during the 1990s through collaborative research projects at BIO and SABS. Geochemical variables in surface sediments (S and Eh potentials) and, for salmon farms, trace metals were found to

We are grateful to H. Akagi, S. Armsworthy, J. Bugden, R. Losier, G. Phillips, and A. Wilson for valuable technical assistance during various research projects on aquaculture-environment interactions conducted at the Bedford Institute of Oceanography and St. Andrews Biological Station over two decades. We thank S. Armstrong, G. Bugden, M. Dowd,

Provision of Advice to Habitat Management

§

§

Voyage of Discovery

page 289

FISHERIES, ECOSYSTEMS AND AQUACULTURE J. Grant, D. Greenberg, G. Harrison, K. Haya, E. Horne, P. Kepkay, W. Li, F. Page, J. Martin, T. Milligan, S. Robinson, W. Silvert, J. Smith, P. Strain, and P. Yeats for their collaboration and participation in field work, data analysis, and modelling. Honours and graduate students L. Auffrey, F. Bouvet, L. Doucette, D. Duplisea, F. Friars, N. Hamilton, and E. Pfeiffer contributed significantly to the projects through their dissertation research. Ron Loucks, Ruth Loucks, and Robin Anderson kindly commented on the manuscript.

References Anderson, J.M. 2007. The Salmon Connection. Glen Margaret Publishing, Tantallon, NS. Anderson, M.R., M.F. Tlusty, and V.A. Pepper. 2005. Organic enrichment at cold water aquaculture sites - the case of coastal Newfoundland. Handbook of Environmental Chemistry 5, Part M: 99-114. Cranford, P.J. and others. 2006. Indicators and thresholds for use in assessing shellfish aquaculture impacts on fish habitat. Canadian Science Advisory Secretariat Research Document 2006/034, Fisheries and Oceans Canada, Ottawa, ON. Cranford, P.J., P.M. Strain, M. Dowd, J. Grant, B.T. Hargrave, and M.-C. Archambault. 2007. Influence of mussel aquaculture on nitrogen dynamics in a nutrient enriched coastal embayment. Marine Ecology Progress Series 347: 61-78. Cranford, P.J., B.T. Hargrave, and L.I. Doucette. 2009. Benthic organic enrichment from suspended mussel (Mytilus edulis) culture in Prince Edward Island, Canada. Aquaculture 292: 189-196. Cranford, P.J., T. Balch, C. Buchan, J. Crocker, R. Filgueira, J. Grant, B.T. Hargrave, B. Law, W. Li, L. Lu, D. Sephton, S. Roach, and R. Stuart. 2011. Mussel aquaculture regulatory effectiveness monitoring: validation of the St. Ann’s Harbour Environmental Assessment and Monitoring Program. Canadian Technical Report of Fisheries Aquatic Sciences 2905, Fisheries and Oceans Canada, Dartmouth, NS. Cranford, P.J., P. Kamermans, G. Krause, A. Bodoy, J. Mazurié, B. Buck, P. Dolmer, D. Fraser, K. Van Nieuwenhove, F.X. O’Beirn, A. Sanchez-Mata, G.G. Thorarinsdóttir, and Ø. Strand. 2012. An ecosystem-based framework for the integrated evaluation and management of bivalve aquaculture impacts. Aquaculture Environment Interactions 2: 193-213. de Margerie, S., A. Isenor, S. Hurlbut, and R. Loucks. 1990. A water quality model for L’Etang Inlet. ASA Consulting, Dartmouth, NS. Fisheries and Oceans Canada. 2003. A scientific review of the potential environmental effects of aquaculture in aquatic ecosystems. Canadian Technical Report of Fisheries Aquatic Sciences 2450, Volume I, Fisheries and Oceans Canada, Ottawa, ON. Fisheries and Oceans Canada. 2004. A scientific review of the potential environmental effects of aquaculture in aquatic ecosystems. Canadian Technical Report of Fisheries Aquatic Sciences 2450, Volume III, Fisheries and Oceans Canada, Ottawa, ON. Fisheries and Oceans Canada. 2006. A scientific review of the potential environmental effects of aquaculture in aquatic ecosystems. Canadian Technical Report of Fisheries Aquatic Sciences 2450, Volume IV, Fisheries and Oceans Canada, Ottawa, ON. Grant, J., C. Bacher, P.J. Cranford, T. Guyondet, and M. Carreau. 2008. A spatially explicit ecosystem model of seston depletion in dense mussel culture. Journal of Marine Systems 73: 155-168. Greenberg, D.A., J.A. Shore, F.H. Page, and M. Dowd. 2005. A finite element circulation model for embayments with drying intertidal areas with application to the Quoddy region of the Bay of Fundy. Ocean Modelling 10: 211-231. Hargrave, B.T. 2002. A traffic light decision system for marine finfish aquaculture siting. Ocean and Coastal Management 45: 215-235. Hargrave, B.T. 2004. Environmental Studies for Sustainable Aquaculture (ESSA): 2004 Symposium Report. Canadian Technical Report of Fisheries Aquatic Sciences 2542, Fisheries and Oceans Canada, Dartmouth, NS. Hargrave, B.T., D.E. Duplisea, E. Pfeiffer, and D.J. Wildish. 1993. Seasonal changes in benthic fluxes of dissolved oxygen and ammonium associated with marine cultured Atlantic salmon. Marine Ecology Progress Series 96: 249-257. Hargrave, B.T., G.A. Phillips, L.I. Doucette, M.J. White, T.G. Milligan, D.J. Wildish, and R.E. Cranston. 1997. Assessing benthic impacts of organic enrichment from marine aquaculture. Water, Air, Soil Pollution 99: 641-650. Hargrave, B.T., L.I. Doucette, P.J. Cranford, B.A. Law, and T.G. Milligan. 2008. Influence of mussel aquaculture on benthic organic enrichment in a nutrientrich coastal embayment. Marine Ecology Progress Series 365: 137-149. Harrison, W.G., T. Perry, and W.K.W. Li. 2005. Ecosystem indicators of water quality. Part I. Plankton biomass, primary production and nutrient demand. Handbook of Environmental Chemistry 5, Part M: 59-82. Kepkay, P.E., W.G. Harrison, and J.B.C. Bugden. 2005. Ecosystem indicators of water quality. Part II. Oxygen production and oxygen demand. Handbook of Environmental Chemistry 5, Part M: 83-97. Loucks, R.H. 1988. Preliminary oceanographic information for salmonid aquaculture in L’Etang Inlet, New Brunswick. Contractor’s Report to DFO, St. Andrews Biological Station, St. Andrews, NB. Loucks, R.H. 1991. Measurement of dispersion in L’Etang Inlet and approaches using drifters. Contractor’s report prepared for New Brunswick Department of Fisheries and Aquaculture, Fredericton, New Brunswick.

§

Voyage of Discovery

§

page 290

Milligan, T.G. and B.A. Law. 2005. The effect of marine aquaculture on fine sediment dynamics in coastal inlets. Handbook of Environmental Chemistry 5, Part M: 239-251. O’Dor, R.K and R. Thompson. 1996. Ocean production enhancement network (OPEN): fourth annual scientific meeting contribution. Marine Biology 124: 591-592. Page, F., D. Greenberg, G. Bugden, R. Losier, J. Shore, E. Horne, S. Robinson, B. Chang, and T. Sephton. 1999. Oceanographic component of the Canadian Department of Fisheries and Oceans Science Strategic Research Program on Coastal Oceanography for sustainable aquaculture development (COSAD). Bulletin of the Aquaculture Association of Canada 99-2: 14-16. Page, F.H., R. Losier, P. McCurdy, D. Greenberg, J. Chaffey, and B. Chang. 2005. Dissolved oxygen and salmon cage culture in the Southwestern New Brunswick Portion of the Bay of Fundy. Handbook of Environmental Chemistry 5, Part M: 1-28. Pohle, G., B. Frost, and R. Findlay. 2001. Assessment of regional benthic impact of salmon mariculture within L’Etang Inlet, Bay of Fundy. ICES Journal of Marine Science 58: 417-426. Robinson, S.M.C., L.M. Auffrey, and M.A. Barbeau. 2005. Far-field impacts of eutrophication on the intertidal zone in the Bay of Fundy, Canada with emphasis on the soft shell clam, Mya arenaria. Handbook of Environmental Chemistry 5, Part M: 253-274. Shaw, J., D.J.W. Piper, G.B.J. Fader, E.L. King, B.J. Todd, T. Bell, M.J. Batterson, and D.G.E. Liverman. 2006. A conceptual model of the deglaciation of Atlantic Canada. Quaternary Science Reviews 25: 2,059-2,081. Silvert, W. 1994. Decision support systems for aquaculture licensing. Journal of Applied Ichthyology 10: 307-311. Silvert, W. and B. Hargrave. 1995. Report on International Workshop on Modelling Environmental Interactions of Mariculture. Proceedings of an ICES Annual Science Conference, C. 1995/F: 6 (Session F), Copenhagen, Denmark. Strain, P.M. and B.T. Hargrave. 2005. Salmon aquaculture, nutrient fluxes and ecosystem processes in southwestern New Brunswick. Handbook of Environmental Chemistry, 5, Part M: 29-57. Strain, P.M., D.J. Wildish, and P.A. Yeats. 1995. The application of simple models of nutrient loading and oxygen demand to the management of a marine tidal inlet. Marine Pollution Bulletin 30: 253-261. Stucchi, D., T.F. Sutherland, C. Levings, and D. Higgs. 2005. Near-field depositional model for salmon aquaculture waste. Handbook of Environmental Chemistry 5, Part M: 157-180. Trites, R.W. 1991. Measurement of dispersion in L’Etang Inlet and approaches using drifters. R.H. Loucks Oceanology Limited., Dartmouth, NS. Trites, R.W. and L. Petrie. 1995. Physical oceanographic features of L’Etang Inlet including evaluation and results from a numerical model. Canadian Technical Report of Hydrography and Ocean Sciences 163, Fisheries and Oceans Canada, Dartmouth, NS. Wildish, D.J. and G.W. Pohle. 2005. Benthic macrofaunal changes resulting from finfish mariculture. Handbook of Environmental Chemistry 5, Part M: 275 – 304. Wildish, D.J., J.L. Martin, A.J. Wilson, and A.M. DeCoste. 1988. Environmental monitoring of the Bay of Fundy salmonid mariculture industry during 1986 and 1987. Canadian Technical Report of Fisheries and Aquatic Sciences 1648, Fisheries and Oceans Canada, St. Andrews, NB. Wildish, D.J., J.L. Martin, R.W. Trites, and A.M. Saulnier. 1990. A proposal for environmental research and monitoring of organic pollution caused by salmonid mariculture in the Bay of Fundy. Canadian Technical Report of Fisheries and Aquatic Sciences 1724, Fisheries and Oceans Canada, St. Andrews, NB. Wildish, D.J., P.D. Keizer, A.J. Wilson, and J.L. Martin. 1993. Seasonal changes of dissolved oxygen and plant nutrients in seawater near salmonid net pens in the macrotidal Bay of Fundy. Canadian Journal of Fisheries and Aquatic Sciences 50: 303-311. Wildish, D.J., H.M. Akagi, N. Hamilton, and B.T. Hargrave. 1999. A recommended method for monitoring sediments to detect organic enrichment from mariculture in the Bay of Fundy. Canadian Technical Report of Fisheries and Aquatic Sciences 2286, Fisheries and Oceans Canada, St. Andrews, NB. Wildish, D.J., G.W. Pohle, B.T. Hargrave, T.F. Sutherland, and M.R. Anderson. 2005. Benthic monitoring methods for habitat management of finfish mariculture in Canada. Canadian Science Advisory Secretariat Research Document 005/039, Fisheries and Oceans Canada, Ottawa, ON. World Wildlife Fund. 2010. Bivalve Aquaculture Sustainability Standards. World Wildlife Fund, Toronto, ON. www.worldwildlife.org/what/globalmarkets/ aquaculture/ WWFBinaryitem17872.pdf: 50 p. Yeats, P.A., T.G. Milligan, T.F. Sutherland, S.M.C. Robinson, J.A. Smith, P. Lawton, and C.D. Levings. 2005. Lithium-normalized zinc and copper concentrations in sediments as measures of trace metal enrichment due to salmon aquaculture. Handbook of Environmental Chemistry 5, Part M: 207-220.